BACKGROUND OF THE INVENTION
[0001] The Global Navigation Satellite System (GNSS), of which one component is the Global
Positioning System (GPS), is a satellite-based navigation system. GNSS permits land,
sea, and airborne users to determine their three-dimensional positions, velocities,
and time. GPS uses NAVSTAR (NAVigation Satellite Timing and Ranging) satellites. The
current satellite constellation consists of 21 operational satellites and 3 active
spares. This constellation provides a GNSS device to receive signals from four to
twelve GNSS satellites at any given time. A minimum of four GNSS satellites allows
the GNSS device to compute its position (latitude, longitude, and altitude) and GNSS
system time. Altitude is typically referenced to mean sea level. The GNSS satellite
signal from the GNSS satellites contains information used to identify the GNSS satellite,
as well as to provide position, timing, ranging data, satellite status, and the updated
ephemeris (orbital parameters). Coarse orbital information is available in the almanac
data. The ephemeris data contains higher accuracy orbital position information.
[0002] Satellite signal acquisition requires a relatively long period of time, particularly
during a cold start initialization of the GNSS device (the GNSS device has no current
location information for itself and/or for any satellites, and/or the GNSS device
does not know time). During warm start initialization, the GNSS device does have some
information, such as relatively recent almanac data and/or ephemeris data, and/or
an estimate of time that may be used to speed up the GNSS satellite signal acquisition
process. However, the time to acquire a GNSS satellite signal for one GNSS satellite,
and the total time required to acquire four or more GNSS satellite signals, is very
noticeable to the user of the GNSS device. That is, the user of the GNSS device has
to wait for some discernable period of time until position information is presented.
Accordingly, it is desirable to reduce the GNSS satellite signal acquisition times.
SUMMARY OF THE INVENTION
[0003] Systems and methods of acquiring Global Navigation Satellite System (GNSS) signal,
such as the Global Positioning System (GPS) signal, are disclosed. An exemplary embodiment
uses satellite orbital position information from almanac and/or ephemeris data to
adjust at least one signal processing parameter, such as reducing the number of analyzed
frequency bins or setting a signal strength threshold, so that GNSS satellite signal
acquisition times are reduced.
[0004] An exemplary embodiment estimates an orbital position for at least one GNSS satellite
based upon at least one of almanac data and ephemeris data, detects a signal emitted
from the at least one GNSS satellite, and based upon the estimated orbital position
information for the at least one GNSS satellite that is determined from the almanac
data and the ephemeris data, adjusts at least one parameter used in the analysis of
the detected signal
[0005] In accordance with further aspects, an exemplary embodiment is an apparatus that
is operable to acquire a GNSS satellite signal comprising a GNSS front end operable
to receive an GNSS signal emitted by at least one GNSS satellite and operable to frequency
downconvert the GNSS signal into a downconverted GNSS satellite signal by mixing the
GNSS satellite signal with a local oscillator signal to shift the carrier frequency
of the incoming GNSS satellite signal and a GNSS system processor. The GNSS system
processor is operable to convert the downconverted GNSS satellite signal to a digital
GNSS satellite signal, process the digital GNSS satellite signal into a plurality
of frequency bins, estimate an orbital position for at least one GNSS satellite based
upon at least one of almanac data and ephemeris data, and based upon the estimated
orbital position information for the at least one GNSS satellite, adjust at least
one signal processing parameter used in a GNSS satellite signal acquisition process,
and based upon the adjusted signal processing parameter, identify at least one of
the frequency bins that corresponds to the GNSS signal emitted by the GNSS satellite.
BRIEF DESCRIPTION OF THE DRAWINGS
[0006] Preferred and alternative embodiments are described in detail below with reference
to the following drawings:
[0007] FIGURE 1 is a simplified block diagram of a Global Navigation Satellite System (GNSS)
device embodiment;
[0008] FIGURE 2 is a block diagram illustrating a GNSS satellite signal correlation process
using an Fast Fourier Transform (FFT) analysis process;
[0009] FIGURE 3 is a conceptually illustrative three dimensional view of a plurality of
frequency bins corresponding to a detected GNSS satellite signal;
[0010] FIGURE 4 is a conceptually illustrative three dimensional view of a reduced number
of frequency bins that are searched during the GNSS satellite signal acquisition process;
[0011] FIGURE 5 is a conceptually illustrative three dimensional view of frequency bins
that are searched during the GNSS satellite signal acquisition process after adjustment
of a signal strength threshold by an embodiment of a GNSS device; and
[0012] FIGURE 6 is a conceptually illustrative three dimensional view of finer resolution
frequency bins that are searched during the GNSS satellite signal acquisition process.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
[0013] Embodiments of a Global Navigation Satellite System (GNSS) device 100, such as the
Global Positioning System (GPS) device, uses estimated satellite orbital position
information obtained from almanac and/or ephemeris data for setting signal strength
threshold, so that GNSS satellite signal acquisition times are reduced. The almanac
and/or ephemeris data may be retrieved from memory or received in a detected satellite
signal. The GNSS device 100 may be interchangeably referred to as a GPS device, satellite
position detection device, or the like.
[0014] FIGURE 1 is a simplified block diagram of a GNSS device embodiment. The GNSS device
100 includes a GNSS front end 102 and a GNSS processor system 104.
[0015] The GNSS front end 102 receives the incoming analog GNSS satellite signal from a
GNSS antenna 106, which detects signals from a plurality of visible GNSS satellites.
The analog GNSS satellite signal is amplified by a low noise amplifier 108 and is
filtered by a bandpass filter 110. The low noise amplifier 108 and the bandpass filter
110 may be provided in multiple stages, if desired.
[0016] The analog GNSS satellite signal at the output of the bandpass filter 110 is downconverted
in a frequency downconverter 112 by mixing the analog GNSS satellite signal with a
local oscillator signal to shift the carrier frequency of the incoming analog GNSS
satellite signal to a lower and more manageable frequency band. This downconversion
can be performed multiple times to bring the frequency of the analog GNSS satellite
signal down in steps to the final desired frequency. Each mixing operation produces
a high frequency information band along with the lower frequency band. Therefore,
each mixing stage of the frequency downconverter 112 may require a bandpass filter
to remove the information in the corresponding higher frequency band. A frequency
reference 114 provides a frequency reference for the frequency downconverter 112.
[0017] The GNSS processor system 104 includes an analog to digital (A/D) sampler 116 which,
in response to the signal from the frequency reference 114, converts the downconverted
analog GNSS satellite signal to a digital GNSS satellite signal. A GNSS processor
118 uses the downconverted digital GNSS signal to first determine ranging and satellite
information for the GNSS satellites in view of its antenna at the time, and to then
determine latitude, longitude, altitude, and/or GNSS system time.
[0018] Acquisition and tracking can be performed by the GNSS processor 118 in software.
Acquiring and tracking the GNSS satellite signals principally means that the code
and frequency offsets between GNSS satellites and GNSS device 100 are determined and
tracked on a regular basis. A typical value of how often the tracking loops execute
is 1000 times per second. GNSS satellite position is determined from these offsets,
but it can be done faster or slower depending on the processing capability available
in the GNSS receiver.
[0019] A GNSS satellite transmits a signal that is detected by the GNSS device 100. The
transmitted GNSS signal from each GNSS satellite is modulated with a psuedo-noise
(PN) code that is a sequence of 1023 chips. Further, the GNSS satellite signal is
modulated by an additional data stream which contains almanac data and the ephemeris
data.
[0020] The PN code in the received GNSS satellite signal is periodically repeated. The GNSS
device 100 generates a plurality of replica PN codes at many different frequencies,
which are compared with the received GNSS satellite signal using a correlation process.
Fast Fourier Transform (FFT) techniques are used to facilitate the correlation process.
An exemplary FFT process is described in
U.S. Patent publication 2007/0046536 to Zhike et. al, entitled "Fast Fourier Transform with Down Sampling Based Navigational Satellite
Signal Tracking", which is incorporated by reference herein. Any suitable GNSS satellite
tracking system that analyzes signals using a plurality of frequency bins may use
embodiments described herein to more quickly and/or more accurately acquire a GNSS
satellite signal.
[0021] The FFT analysis process correlates the GNSS satellite signal information to generate
correlation values that are stored into a series of frequency bins over a time range.
The GNSS satellite signal will have its unique PN code corresponding to at least one
particular frequency bin. Thus, the PN signal corresponds to at least one of the frequency
bins that are searched by the GNSS device 100 during the GNSS signal acquisition process.
The PN code may be detectable over a relatively small frequency range, and thus, the
GNSS satellite signal may be detectable in several adjacent frequency bins depending
upon the frequency range of the frequency bins defined during the FFT analysis process.
[0022] Once the GNSS signal is "acquired" (when the frequency and time of the GNSS satellite
signal is identified by the correlation process), the GNSS processor 28 may determine
its relative distance to the GNSS satellite. When at least four GNSS satellite signals
have been acquired, and the corresponding GNSS satellite orbital positions determined,
a least squared error triangulation process is used to determine the location of the
GNSS device 100 relative to the GNSS satellites (GNSS position).
[0023] Since the position of the GNSS satellites are known relative to the earth, the GNSS
processor 118 determines latitude and longitude of the GNSS device 100. The GNSS processor
118 can also determine altitude (or depth) if the GNSS device 100 operates in three
dimensional space. The GNSS processor 118 may further determine rate and GNSS system
time from the acquired GNSS signal.
[0024] As noted above, the GNSS data in a received GNSS satellite signal is processed by
the GNSS processor 118 in the frequency domain (instead of the time domain). Frequency
domain processing techniques use FFT for Wavelet Multiresolution Analysis (WMA). Such
software based GNSS signal acquisition may be faster than hardware based signal acquisition,
and software processing can eliminate much of the front end processing. However, while
frequency domain techniques used during signal acquisition are faster than conventional
time domain techniques, there will still be a discernable amount of time required
to complete the computations required for GNSS satellite signal acquisition. Thus,
a delay can occur from the end of the batch data collection and the completion of
the signal acquisition phase depending on the processor 118 speed.
[0025] The GNSS system processor 118 further includes a memory 120. As noted above, the
received GNSS satellite signal includes almanac data and ephemeris data. The information
in the received almanac data and/or ephemeris data is saved into the almanac database
122 and/or ephemeris database 124 portions of memory 120 in a suitable format, such
as in a database or table. Coarse satellite orbital position information for a plurality
of GNSS satellites resides in the almanac database 122. More accurate satellite orbital
position information resides in the ephemeris database 124. Other embodiments may
store the received almanac data and/or ephemeris data in other formats, memory regions,
or even in other memory devices.
[0026] The GNSS estimation logic 126 is software stored in memory 120 that is used to estimate
GNSS satellite positions. One skilled in the art will appreciate that the GNSS estimation
logic 126 can be stored on any computer-readable medium for use by or in connection
with any computer and/or processor related system or method. The GNSS position estimation
logic 126 (retrieved and executed by the GNSS processor 28) estimates orbital position
of selected visible GNSS satellites based upon the satellite orbital position information
in the almanac database 122 and/or the ephemeris database 124.
[0027] Almanac data and ephemeris data is transmitted in portions in the GNSS satellite
signals, and accordingly, some period of time is required to receive a complete set
of almanac data and ephemeris data. Relatively recently acquired almanac data and
ephemeris data is saved into the almanac database 122 and the ephemeris database 124,
respectively, of memory 120. The receipt dates and times of the almanac and ephemeras
data is also saved such that as assessment of the validity of the almanac data and
the ephemeris data may be made by the GNSS processor 118 during the GNSS signal acquisition
process. That is, the the almanac database 122 and/or the ephemeris database 124 must
be relatively current for estimation of GNSS satellite orbital positions relative
to the location of the GNSS device 100. Any suitable process for estimating the orbital
position of a GNSS satellite relative to the GNSS device 100 may be used by the various
embodiments of the GNSS device 100.
[0028] Accordingly, if information in the almanac database 122 and/or the ephemeris database
124 is to be used to estimate GNSS satellite orbital positions with a suitable degree
of accuracy, then the information in the almanac database 122 and/or the ephemeris
database 124 must be relatively current. In one embodiment, a coarse satellite orbital
position is estimated based upon the almanac information in the almanac database 122.
Alternatively, a higher accuracy satellite orbital position may be estimated based
upon the information in the ephemeris database 124. An estimate of current time may
be required for either estimation.
[0029] However, if the information in the almanac database 122 and/or the ephemeris database
124 is not current, and/or an estimate of current time is not available, the estimated
orbital positions of the GNSS satellites may be relatively inaccurate such that the
signal acquisition process will not be able to utilize features of the various embodiments
described herein. In such situations, the GNSS satellite signal acquisition process
is performed using legacy GNSS signal acquisition techniques.
[0030] Once the GNSS processor 118 has estimated the orbital position for at least one visible
GNSS satellite, embodiments of the GNSS device 100 use the estimated satellite orbital
position information to adjust at least one parameter used in the FFT analysis of
the detected GNSS satellite signal. One or more signal processing parameters may be
adjusted by the various embodiments of the GNSS device 100.
[0031] Additionally, or alternatively, some embodiments of the GNSS device 100 use a signal
threshold 128 parameter. The signal threshold 128 is a predefined parameter that is
initialized during startup of the GNSS device 100. The signal threshold 128 is used
to distinguish incoming weak signals, and/or signals with high levels of noise, from
an incoming GNSS satellite signal.
[0032] In the various embodiments, the signal threshold 128 is an adjustable signal power
threshold value, and/or is an adjustable signal-to-noise threshold value. Based upon
the estimated orbital position of the GNSS satellite, the received signal strength
and/or signal-to-noise (SN) ratio may be estimated for a signal emitted by the GNSS
satellite. For example, a signal emitted from a GNSS satellite in an orbital position
near the horizon will be significantly attenuated and/or distorted by the earth's
atmosphere. On the other hand, a signal emitted from a GNSS satellite in an orbital
position that is substantially overhead will be less attenuated and/or distorted by
the earth's atmosphere.
[0033] Based on the estimated power and/or SN ratio, GNSS satellite signals that have characteristics
corresponding to the estimated power and/or SN ratio can be readily identified. For
example, in a plurality of frequency bins generated by the FFT analysis process, a
frequency bin corresponding to the GNSS satellite will have a discernable correlation
peak (a relatively high acquisition value, for example). Such frequency bins can be
identified by their correlation peaks. Thus, an estimate is made to determine how
much taller the correlation peak should be over the expected noise.
[0034] The signal power threshold and/or SN ratio threshold may be adjusted based on an
estimated signal strength or an expected SN ratio that is determined from the estimated
orbital position of the GNSS satellite. Accordingly, frequency bins having a corresponding
signal strength greater than the adjusted signal strength threshold, or having a SN
ratio greater than the SN ratio threshold, are identified during the GNSS satellite
signal acquisition process. The identified frequency bins are then used for the GNSS
satellite signal acquisition process.
[0035] FIGURE 2 is a block diagram illustrating a GNSS satellite signal correlation process
using an FFT analysis process. The incoming digitized and downconverted GNSS satellite
signal [r(n)] is processed by the GNSS system processor 104 (FIGURE 1) using FFT techniques
over a range of frequencies (f
1, f
2, .... f
k). At each particular frequency, the GNSS satellite signal r(n) is multiplied by cosine
and sine functions. An FFT is computed of that mixed signal. The resultant FFT is
then multiplied by the complex conjugate of the replica code for the satellite being
searched at each frequency being searched. This previously determined complex conjugate
FFT is retrieved from a look-up table or the like. The resulting complex conjugate
is further processed by computing the inverse FFT (iFFT) to derive a magnitude value
for each of the selected frequencies. The value is then stored in a frequency bin.
[0036] FIGURE 3 is a conceptually illustrative three dimensional view of a plurality of
frequency bins 300 corresponding to a detected GNSS satellite signal. Most of the
frequency bins 302 have a relatively low numerical value of the acquisition ratio.
The numerical values of the acquisition ratio correspond to the degree of correlation
between the signal hypothesis generated by the GNSS device 100 and the received GNSS
satellite signal. The low numerical values of the acquisition ratio indicates detected
noise or the like in the GNSS satellite signal for that particular frequency bin.
[0037] However, some of the frequency bins 304 have a very high value for its acquisition
ratio. Here, the high value of the acquisition ratio corresponds to those frequency
bins 304 that identify the frequency and time of the detected GNSS satellite signal.
Legacy GNSS devices search the entire set of frequency bins 300 looking for the frequency
bins 304. It is appreciated that such a search process, which unnecessarily includes
processing of information in the frequency bins 302, is very time consuming.
[0038] One adjustable parameter is the number of frequency bins (and/or their respective
location in a matrix) processed during the GNSS satellite signal acquisition process.
FIGURE 4 is a conceptually illustrative three dimensional view of a reduced number
of frequency bins that are searched during the GNSS satellite signal acquisition process
performed by an embodiment of the GNSS device 100. For example, one embodiment deselects
some or all of the frequency bins 302 that are not expected to contain information
corresponding to the GNSS satellite signal. For example, identified ones of the frequencies
f
1 to f
k (FIGURE 2) may be deselected or the like.
[0039] Frequency bins that are deselected are identified based upon the estimated location
of the GNSS satellite. The entire range of searchable frequency bins correspond to
a range of possible locations of the GNSS satellite that is being acquired. For example,
if the almanac data and/or ephemeris data is used to estimate that the GNSS satellite
is directly overhead, it is appreciated that frequency bins corresponding to positions
other than substantially overhead do not need to be searched. Thus, embodiments of
the GNSS device 100 deselect those frequency bins that do not correspond to a substantially
overhead position.
[0040] One embodiment estimates location of the GNSS satellite based upon the almanac data
and/or ephemeris data. The embodiment identifies a predefined range of frequency bins
about the frequency bin(s) that corresponds to the estimated location. Bins outside
of that range ore deselected.
[0041] Thus, frequency bins are screened out, deleted, or otherwise identified as frequency
bins that do not need to be processed during the GNSS satellite signal acquisition
process. Accordingly, a reduced number of frequency bins 402 are processed by the
GNSS device 100 during the GNSS satellite signal acquisition process. Another embodiment
may select frequencies for a plurality of frequency bins that are expected to include
the frequency bins 304. For example, selected ones of the frequencies f
1 to f
k (FIGURE 2) may be selected. Thus, signal acquisition frequency bins are selected
for processing during the GNSS satellite signal acquisition process.
[0042] In another embodiment, the above-described signal threshold 128 (FIGURE 1) may be
adjusted based upon the expected characteristics of the GNSS satellite signal. FIGURE
5 is a conceptually illustrative three dimensional view of frequency bins that are
searched during the GNSS satellite signal acquisition process after adjustment of
a signal strength threshold 502 by an embodiment of a GNSS device 100. The signal
strength threshold 502 may be used to effectively screen out signals having signal
strengths less than the threshold. As noted above, frequency bins which correspond
to the acquired GNSS satellite signal have a discernable numerical value of the acquisition
ratio. Other frequency bins that do not correspond to the GNSS satellite signal have
no, or relatively small, numerical values for its acquisition ratio and can be identified.
Such frequency bins are screened out, deleted, or otherwise identified as frequency
bins that do not need to be processed during the GNSS satellite signal acquisition
process.
[0043] In the exemplary frequency bins illustrated in FIGURE 5, the signal strength threshold
502 has been adjusted such that frequency bins 504 having information corresponding
to a signal strength less than the signal strength threshold 502 have been screened
out, deleted, or otherwise identified as frequency bins that do not need to be processed
during the GNSS satellite signal acquisition process. Some frequency bins 506 may
have corresponding numerical values of their acquisition ratio that exceeds the signal
strength threshold 502. These frequency bins 506 are selected for processing. In one
embodiment, other screening processes or techniques may be used to screen out, delete,
or otherwise identify some of these remaining frequency bins 506 as frequency bins
that do not need to be further processed during the GNSS satellite signal acquisition
process.
[0044] The frequency bins 304 corresponding to the GNSS satellite signal are identifiable
by their acquisition ratio numerical values which correspond to having a signal strength
that is greater than or equal to the signal strength threshold 502. Accordingly, the
frequency bins 304 are readily identifiable. These frequency bins 304 may be selected
for processing by the GNSS device 100 during the GNSS satellite signal acquisition
process.
[0045] FIGURE 6 is a conceptually illustrative three dimensional view of finer resolution
frequency bins 602 that are searched during the GNSS satellite signal acquisition
process. When orbital position of a GNSS satellite is estimated using the almanac
database 122 and/or the ephemeris database 124, as noted above, frequency bins that
are expected to contain information corresponding to the GNSS satellite signal may
be identified. Such bins are identifiable by their time and by frequency. An alternative
embodiment of the GNSS device 100 uses this information to construct higher resolution
frequency bins 602. That is, the number of frequency bins 602 that are processed remains
the same, or at least remain at a relatively large number compared to the number of
bins 402 (FIGURE 4). However, each frequency bin 602 corresponds to a narrower time
period and/or a narrower frequency range. For example, the total frequency range of
the processed frequency bins 602 is from 1.0415 x 10
7 Hz to 1.0420 x 10
7 Hz. In contrast, the the total frequency range of the processed frequency bins 300
(FIGURE 3) is from 1.0415 x 10
7 Hz to 1.0425 x 10
7 Hz. Thus, the frequency range of each frequency bin 602 is half of the frequency
range of the frequency bins 300 (FIGURE 3).
[0046] Adjusting the time period and/or a frequency range so that the frequency bins have
a higher resolution allows identification of the GNSS satellite signal with greater
precision. Thus, the satellite's orbital position can be determined with a greater
accuracy. Further, during the subsequent tracking process, the GNSS satellite signal
may be more accurately tracked since its signal has been located with a higher degree
of accuracy by using the higher resolution frequency bins 602 that correspond to a
smaller time period and/or frequency range.
[0047] The embodiments of the invention in which an exclusive property or privilege is claimed
are defined as follows:
1. A method for acquiring a Global Navigation Satellite System (GNSS) signal, the method
comprising:
estimating an orbital position for at least one GNSS satellite based upon at least
one of almanac data and ephemeris data;
detecting a signal emitted from the at least one GNSS satellite; and
based upon the estimated orbital position information for the at least one GNSS satellite,
adjusting at least one signal processing parameter used in a GNSS satellite signal
acquisition process.
2. The method of Claim 1, wherein adjusting at least one signal processing parameter
comprises:
based upon the estimated orbital position information for the at least one GNSS satellite,
estimating a characteristic of the detected signal;
adjusting a signal threshold based upon the estimated characteristic; and
identifying at least one frequency bin having a corresponding signal characteristic
that corresponds to the adjusted signal threshold such that the identified frequency
bin is used in the GNSS satellite signal acquisition process.
3. The method of Claim 1, wherein adjusting at least one signal processing parameter
comprises:
based upon the estimated orbital position information for the at least one GNSS satellite,
estimating a signal-to-noise ratio of a signal emitted by the at least one GNSS satellite;
adjusting a signal-to-noise ratio threshold based upon the estimated signal-to-noise
ratio; and
identifying at least one frequency bin having a corresponding signal-to-noise ratio
that is greater than the adjusted signal-to-noise ratio threshold such that the identified
at least one frequency bin is used in the GNSS satellite signal acquisition process.
4. The method of Claim 1, wherein adjusting at least one signal processing parameter
comprises:
based upon the estimated orbital position information for the at least one GNSS satellite,
estimating a signal strength of the detected signal;
adjusting a signal strength threshold based upon the estimated signal strength; and
identifying at least one frequency bin having a corresponding signal strength that
corresponds to the adjusted signal strength threshold such that the identified frequency
bin is used in the GNSS satellite signal acquisition process.
estimating a coarse satellite orbital position based upon the almanac data.
estimating a high accuracy satellite orbital position based upon the ephemeris data.
5. The method of Claim 1, wherein estimating the orbital position for the at least one
GNSS satellite based upon at least one of the almanac data and the ephemeris data
comprises;
estimating a coarse satellite orbital position based upon the almanac data; and
after estimation of the coarse satellite orbital position, estimating a high accuracy
satellite orbital position based upon the ephemeris data.
6. The method of Claim 1, wherein adjusting at least one parameter comprises:
adjusting a frequency range of a frequency bin based upon the estimated orbital position
information for the at least one GNSS satellite.
7. The method of Claim 1, wherein adjusting at least one parameter comprises:
adjusting a time range of a frequency bin based upon the estimated orbital position
information for the at least one GNSS satellite.
8. An apparatus that is operable to acquire a Global Positioning System (GNSS) signal,
comprising:
a GNSS front end (102) operable to receive a GNSS signal emitted by at least one GNSS
satellite and operable to frequency downconvert the GNSS signal into a downconverted
GNSS satellite signal by mixing the GNSS satellite signal with a local oscillator
signal to shift the carrier frequency of the incoming GNSS satellite signal; and
a GNSS system processor (104) communicatively coupled to the GNSS front end and operable
to:
receive the downconverted GNSS satellite signal from the GNSS front end;
convert the downconverted GNSS satellite signal to a digital GNSS satellite signal;
process the digital GNSS satellite signal into a plurality of frequency bins;
estimate an orbital position for at least one GNSS satellite based upon at least one
of almanac data and ephemeris data;
based upon the estimated orbital position information for the at least one GNSS satellite,
adjust at least one signal processing parameter used in a GNSS satellite signal acquisition
process; and
based upon the adjusted signal processing parameter, identify at least one of the
frequency bins that corresponds to the GNSS signal emitted by the GNSS satellite.
9. The apparatus of Claim 8, wherein based upon the adjusted signal processing parameter,
the processing system (104) is operable to:
estimate a signal-to-noise ratio corresponding to the received GNSS signal emitted
by the at least one GNSS satellite;
adjust a signal-to-noise ratio threshold based upon the estimated signal-to-noise
ratio; and
identify at least one frequency bin having a corresponding signal-to-noise ratio that
is greater than the adjusted signal-to-noise ratio threshold such that the identified
at least one frequency bin is used in the GNSS satellite signal acquisition process.
10. The apparatus of Claim 8, wherein based upon the adjusted signal processing parameter,
the processing system (104) is operable to:
estimate a signal strength corresponding to the received GNSS signal emitted by the
at least one GNSS satellite;
adjust a signal strength threshold based upon the estimated signal strength; and
identify at least one frequency bin having a corresponding signal strength that corresponds
to the adjusted signal strength threshold such that the identified frequency bin is
used in the GNSS satellite signal acquisition process.